pdf -> jpeg mit convert/popen

1 files changed, 22 insertions, 0 deletions

diff --git a/web.py b/web.py
index 577e56f..dc1632e 100644
--- a/web.py
+++ b/web.py
@@ -13,6 +13,7 @@ import whoosh.query as query
 import flask
 import pyPdf
 import werkzeug
+import subprocess
 
 app = flask.Flask("booksearch")
 
@@ -31,6 +32,27 @@ def do_download_file(docnum):
         filename=os.path.basename(document['path']))
     return r
 
+@app.route("/test")
+def test():
+    input = pyPdf.PdfFileReader(file("test.pdf", 'r'))
+    page = input.getPage(0)
+    out = pyPdf.PdfFileWriter()
+    out.addPage(page)
+    pdffile = StringIO()
+    out.write(pdffile)
+    process = subprocess.Popen(["/usr/bin/convert", "pdf:-", "jpeg:-"], stdin=subprocess.PIPE, stdout=subprocess.PIPE)
+    outjpg = StringIO()
+    pdffile.seek(0)
+    while not process.stdin.closed:
+        print "."
+        stdout, stderr = process.communicate(pdffile.read())
+        if not stdout:
+            break
+        outjpg.write(stdout)
+    outjpg.seek(0)
+    r = werkzeug.Response(outjpg, mimetype="image/jpeg")
+    return r
+
 @app.route("/download/page/<int:docnum>/<int:page>", methods=["GET"])
 def do_download_page(docnum,page):
     document = index.reader().stored_fields(docnum)